Embodiments of the disclosure compensate for global movement and in-scene movement during image capture by a computing device. A sequence of images is accessed by the computing device. Accelerometer readings and/or gyroscope readings corresponding to each of the accessed images are used by the computing device for calculating global movement among each of the accessed images. Each of the accessed images is re-aligned based on the calculated global movement. The re-aligned images are combined into a single output image. The intensity values of each of the pixels in the re-aligned images are compared with the intensity values of each of the corresponding pixels in a reference image. Based on the comparison, the intensity values associated with the pixels in the re-aligned images are selectively accumulated to generate an output image that is blur-free, low-light enhanced, and high dynamic range.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A system for compensating for in-scene and camera motion during image capture by a mobile computing device, said system comprising: an image sensor; a gyroscope for providing rotational motion data of the image sensor; an accelerometer for providing linear motion data of the image sensor; a memory area storing a plurality of images captured by the image sensor, the memory area further storing readings from the gyroscope and the accelerometer, the readings corresponding to each of the plurality of images; and a processor programmed to: calculate global movement among the plurality of images based on the readings, the global movement comprising the rotational motion data and the linear motion data; re-align each of the plurality of images based on the calculated global movement; compare intensity values associated with pixels in the re-aligned images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image in the plurality of images; selectively accumulate the intensity values associated with the pixels in the re-aligned images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and generate an output image using the selectively accumulated intensity values.
A mobile device captures blur-free, low-light, high dynamic range images by compensating for both camera and in-scene movement. It uses an image sensor to capture multiple images and a gyroscope/accelerometer to record rotational and linear motion during capture. The device calculates global movement between images based on the gyroscope/accelerometer data. It then realigns the images to correct for this movement. Finally, the device compares each pixel's intensity in the realigned images against the corresponding pixel in a reference image. Pixel intensity values are selectively accumulated: values significantly above or below the reference pixel's intensity are discarded to reduce blur, while valid intensity values are added to generate the final output image.
2. The system of claim 1 , wherein calculating the global movement among the plurality of images comprises applying a filter on the readings to calculate the global movement among the plurality of images.
This invention relates to image processing systems that analyze movement across multiple images, particularly for applications like video stabilization or motion tracking. The problem addressed is accurately determining global movement (e.g., camera shake or object motion) in a sequence of images, which is often corrupted by noise or local inconsistencies. The system processes a plurality of images to calculate global movement by applying a filter to the motion readings derived from the images. The filter smooths or refines the raw motion data to isolate the true global movement signal from noise or local artifacts. This filtered calculation improves the accuracy of motion estimation, enabling more reliable stabilization or tracking in subsequent applications. The system may also include components for capturing or receiving the images, extracting motion readings (e.g., optical flow or feature tracking), and applying the filter to these readings. The filter could be a low-pass filter, median filter, or other noise-reduction technique tailored to the specific motion characteristics of the images. The filtered output provides a refined global movement estimate, which can then be used for tasks like image alignment, stabilization, or motion compensation. This approach enhances motion estimation by mitigating the effects of noise and local distortions, resulting in more stable and accurate motion analysis.
3. The system of claim 1 , wherein the memory area further stores two frame buffers, wherein a first one of the frame buffers stores each one of the plurality of images as the plurality of images are captured in sequence, and wherein a second one of the frame buffers stores the reference image.
This invention relates to a system for image processing, specifically for capturing and comparing a sequence of images to a reference image. The system includes a memory area that stores two frame buffers. The first frame buffer sequentially stores each image as they are captured in real-time, while the second frame buffer retains a reference image for comparison purposes. This dual-buffer approach allows for continuous image acquisition without overwriting the reference image, enabling real-time analysis or monitoring applications. The system likely addresses the need for efficient image storage and retrieval in applications where a reference image must be preserved while new images are continuously captured, such as in surveillance, quality control, or medical imaging. The use of separate buffers ensures that the reference image remains unchanged while the system processes incoming images, improving accuracy and reliability in comparative analysis. The invention may be part of a larger imaging system that includes a camera or sensor for capturing images and a processor for analyzing the stored data.
4. The system of claim 1 , wherein the selective accumulating further comprises accumulating intensity values associated with the pixels in the re-aligned images falling within a range of intensity values associated with pixels in the reference image and discarding the intensity values associated with the pixels in the re-aligned images falling outside the range of intensity values associated with the pixels in the reference image, the range being between the first threshold and the second threshold.
This invention relates to image processing systems that enhance image quality by selectively accumulating intensity values from multiple re-aligned images. The problem addressed is the presence of noise and artifacts in images captured under varying conditions, such as low light or motion blur, which degrade visual clarity. The system improves image quality by combining multiple re-aligned images while selectively accumulating pixel intensity values based on their similarity to a reference image. The system processes a set of re-aligned images, each representing the same scene but captured under different conditions. A reference image is selected, and its pixel intensity values define a range between a first and second threshold. During accumulation, intensity values from the re-aligned images are compared to this range. Values within the range are retained, while those outside are discarded. This selective accumulation reduces noise and artifacts by prioritizing pixels that closely match the reference image, resulting in a higher-quality final image. The method ensures that only relevant intensity values contribute to the final output, enhancing clarity and reducing distortions. The system is particularly useful in applications requiring high-precision imaging, such as medical imaging, surveillance, and scientific research.
5. The system of claim 1 , wherein the processor is further programmed to capture the plurality of images from the image sensor in sequence and obtain the readings from the gyroscope and the accelerometer simultaneously therewith.
The image stabilization system captures images from the image sensor and simultaneously obtains gyroscope and accelerometer readings. This synchronization ensures that the motion data accurately reflects the camera movement during the exact moment each image was captured, critical for precise global movement calculation and subsequent image realignment. The system is "The system for compensating for in-scene and camera motion during image capture by a mobile computing device, said system comprising: an image sensor; a gyroscope for providing rotational motion data of the image sensor; an accelerometer for providing linear motion data of the image sensor; a memory area storing a plurality of images captured by the image sensor, the memory area further storing readings from the gyroscope and the accelerometer, the readings corresponding to each of the plurality of images; and a processor programmed to: calculate global movement among the plurality of images based on the readings, the global movement comprising the rotational motion data and the linear motion data; re-align each of the plurality of images based on the calculated global movement; compare intensity values associated with pixels in the re-aligned images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image in the plurality of images; selectively accumulate the intensity values associated with the pixels in the re-aligned images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and generate an output image using the selectively accumulated intensity values."
6. The system of claim 1 , wherein the first threshold and the second threshold are configurable.
A system for managing configurable thresholds in a technical process involves adjusting operational parameters to optimize performance. The system includes a monitoring module that tracks real-time data from a process, such as temperature, pressure, or signal strength, and compares it against predefined thresholds. These thresholds define acceptable operating ranges, triggering alerts or adjustments when exceeded. The first threshold represents a lower boundary, while the second threshold represents an upper boundary. Both thresholds are configurable, allowing users to modify them based on changing conditions or requirements. The system dynamically adjusts these thresholds to maintain efficiency, prevent failures, or adapt to varying environmental factors. By enabling customization, the system ensures flexibility in different operational scenarios, improving reliability and performance across diverse applications. The configurable nature of the thresholds allows for fine-tuning without requiring hardware changes, reducing costs and downtime. This approach is particularly useful in industries like manufacturing, energy, or telecommunications, where precise control over operational limits is critical. The system may also include a user interface for setting and modifying thresholds, ensuring ease of use and adaptability.
7. A method for compensating for in-scene and camera motion during image capture, the method comprising: accessing images captured by a computing device; accessing readings from a gyroscope and one or more accelerometers associated with the computing device, each of the readings corresponding to at least one of the accessed images; calculating global movement among each of the accessed images based on the accessed readings, calculating global movement including calculating linear movements via readings from the one or more accelerometers and calculating rotational movements via readings from the gyroscope; re-aligning each of the accessed images based on the calculated global movement; comparing intensity values associated with pixels in the re-aligned accessed images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image selected from the accessed images; selectively accumulating the intensity values associated with the pixels in the re-aligned accessed images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned accessed images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and combining the re-aligned accessed images into a single output image.
A method for compensating for motion during image capture involves accessing multiple images captured by a device and their corresponding gyroscope/accelerometer readings. Global movement between images is calculated using this motion data. The images are realigned based on calculated movement. Each pixel's intensity is compared to its expected value derived from a reference image. Intensity values that are much higher or lower than the expected value (outside a defined threshold range) are discarded to reduce blur. The remaining intensity values are accumulated, creating a cumulative intensity value for each pixel. These accumulated values are then used to generate a single, stabilized output image.
8. The method of claim 7 , wherein combining the re-aligned accessed images comprises combining intensity values from corresponding pixels from each of the re-aligned accessed images to produce intensity values associated with the single output image, while preventing saturation of the single output image.
The method for compensating for in-scene and camera motion during image capture, where the method comprising: accessing images captured by a computing device; accessing readings from a gyroscope and one or more accelerometers associated with the computing device, each of the readings corresponding to at least one of the accessed images; calculating global movement among each of the accessed images based on the accessed readings, calculating global movement including calculating linear movements via readings from the one or more accelerometers and calculating rotational movements via readings from the gyroscope; re-aligning each of the accessed images based on the calculated global movement; comparing intensity values associated with pixels in the re-aligned accessed images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image selected from the accessed images; selectively accumulating the intensity values associated with the pixels in the re-aligned accessed images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned accessed images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and combining the re-aligned accessed images into a single output image, combines the re-aligned images by combining intensity values from corresponding pixels, while also preventing the intensity values from exceeding the maximum allowed value, which would lead to saturation in the output image.
9. The method of claim 7 , wherein the accessed images each include a plurality of pixels, the method further comprising defining a plurality of superpixels each corresponding to a predefined number of pixels derived from the plurality of pixels in each of the accessed images, and wherein said calculating, said re-aligning, and said combining operate on the defined superpixels.
The method for compensating for in-scene and camera motion during image capture, where the method comprising: accessing images captured by a computing device; accessing readings from a gyroscope and one or more accelerometers associated with the computing device, each of the readings corresponding to at least one of the accessed images; calculating global movement among each of the accessed images based on the accessed readings, calculating global movement including calculating linear movements via readings from the one or more accelerometers and calculating rotational movements via readings from the gyroscope; re-aligning each of the accessed images based on the calculated global movement; comparing intensity values associated with pixels in the re-aligned accessed images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image selected from the accessed images; selectively accumulating the intensity values associated with the pixels in the re-aligned accessed images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned accessed images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and combining the re-aligned accessed images into a single output image, defines superpixels which are each comprised of a group of pixels within the image, and performs the movement calculation, realignment, and combining processes on these superpixels rather than on individual pixels.
10. The method of claim 7 , wherein said calculating and said re-aligning occur without operating mechanical parts within the computing device.
The method for compensating for in-scene and camera motion during image capture, where the method comprising: accessing images captured by a computing device; accessing readings from a gyroscope and one or more accelerometers associated with the computing device, each of the readings corresponding to at least one of the accessed images; calculating global movement among each of the accessed images based on the accessed readings, calculating global movement including calculating linear movements via readings from the one or more accelerometers and calculating rotational movements via readings from the gyroscope; re-aligning each of the accessed images based on the calculated global movement; comparing intensity values associated with pixels in the re-aligned accessed images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image selected from the accessed images; selectively accumulating the intensity values associated with the pixels in the re-aligned accessed images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned accessed images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and combining the re-aligned accessed images into a single output image, calculates and realigns the images using software algorithms without physically moving any parts inside the computing device.
11. The method of claim 7 , wherein accessing the images comprises accessing a plurality of images in sequence with one of the plurality of images representing the reference image.
The method for compensating for in-scene and camera motion during image capture, where the method comprising: accessing images captured by a computing device; accessing readings from a gyroscope and one or more accelerometers associated with the computing device, each of the readings corresponding to at least one of the accessed images; calculating global movement among each of the accessed images based on the accessed readings, calculating global movement including calculating linear movements via readings from the one or more accelerometers and calculating rotational movements via readings from the gyroscope; re-aligning each of the accessed images based on the calculated global movement; comparing intensity values associated with pixels in the re-aligned accessed images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image selected from the accessed images; selectively accumulating the intensity values associated with the pixels in the re-aligned accessed images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned accessed images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and combining the re-aligned accessed images into a single output image, accesses a sequence of multiple images, and uses one of those images as the reference image for comparison.
12. The method of claim 11 , wherein calculating the global movement comprises calculating pixel offset between each of the plurality of images and the reference image, the calculated pixel offset including a two-dimensional pixel shift with accounting for rotation of the computing device.
The method for compensating for in-scene and camera motion during image capture, where the method comprising: accessing images captured by a computing device; accessing readings from a gyroscope and one or more accelerometers associated with the computing device, each of the readings corresponding to at least one of the accessed images; calculating global movement among each of the accessed images based on the accessed readings, calculating global movement including calculating linear movements via readings from the one or more accelerometers and calculating rotational movements via readings from the gyroscope; re-aligning each of the accessed images based on the calculated global movement; comparing intensity values associated with pixels in the re-aligned accessed images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image selected from the accessed images; selectively accumulating the intensity values associated with the pixels in the re-aligned accessed images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned accessed images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and combining the re-aligned accessed images into a single output image where accessing the images comprises accessing a plurality of images in sequence with one of the plurality of images representing the reference image, calculates movement by determining the pixel offset between each image and the reference image, factoring in both the 2D shift and any rotation of the device.
13. The method of claim 12 , wherein re-aligning each of the accessed images comprises adjusting pixel locations in each of the plurality of images based on the calculated pixel offset for that image.
This invention relates to image processing, specifically methods for aligning multiple images to correct misalignment caused by movement or distortion. The problem addressed is the need to accurately re-align images in a sequence or set where individual frames or images have shifted pixel positions due to factors like camera motion, environmental changes, or sensor inaccuracies. Misaligned images can degrade the quality of subsequent processing, such as stitching, stacking, or analysis. The method involves accessing a plurality of images that require alignment, where each image has a calculated pixel offset indicating its misalignment relative to a reference image or a common coordinate system. The re-alignment process adjusts the pixel locations in each image based on its specific pixel offset. This adjustment corrects the positional discrepancies, ensuring that the images are properly aligned for further processing. The method may also involve determining the pixel offsets through techniques such as feature matching, cross-correlation, or optical flow analysis, though these steps are not explicitly detailed in the claim. By re-aligning the images, the invention improves the accuracy of image-based applications, such as panoramic stitching, high-dynamic-range imaging, or medical imaging, where precise alignment is critical. The adjustment of pixel locations ensures that the final output is free from artifacts caused by misalignment, enhancing overall image quality and usability.
14. The method of claim 7 , wherein the first threshold and the second threshold are configurable.
The method for compensating for in-scene and camera motion during image capture, where the method comprising: accessing images captured by a computing device; accessing readings from a gyroscope and one or more accelerometers associated with the computing device, each of the readings corresponding to at least one of the accessed images; calculating global movement among each of the accessed images based on the accessed readings, calculating global movement including calculating linear movements via readings from the one or more accelerometers and calculating rotational movements via readings from the gyroscope; re-aligning each of the accessed images based on the calculated global movement; comparing intensity values associated with pixels in the re-aligned accessed images to corresponding expected intensity values for the pixels, the corresponding expected intensity values being derived from a reference image selected from the accessed images; selectively accumulating the intensity values associated with the pixels in the re-aligned accessed images based on the comparison, the selective accumulating comprising, for each pixel in the re-aligned accessed images, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to the corresponding expected intensity value and producing a cumulative intensity value for that pixel; and combining the re-aligned accessed images into a single output image, allows you to change the values of the upper and lower thresholds used to determine which pixel values are discarded during the accumulation process.
15. One or more computer memories embodying computer-executable components, said components comprising: a memory component that when executed causes at least one processor to access a first image and a second image from a sequence of images captured by a computing device; a threshold component that when executed causes at least one processor to compare intensity values associated with pixels in the second image to intensity values associated with corresponding pixels in the first image; a cumulative component that when executed causes at least one processor to compensate for at least one of subject movement or movement of the computing device by selectively accumulating the intensity values associated with the pixels in the first image with the intensity values associated with the corresponding pixels in the second image based on the comparison performed by the threshold component to produce cumulative intensity values, the selective accumulating comprising, for each pixel in the second image, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to corresponding intensity value in the first image and producing a cumulative intensity value for that pixel; and a display component that when executed causes at least one processor to produce an output image having the selectively accumulated intensity values.
Software components stabilize images by accessing a sequence of captured images. A threshold component compares pixel intensities in a second image to corresponding pixels in a first image (which will become the refernce image). A cumulative component selectively accumulates intensity values, compensating for subject or device movement. For each pixel, if its intensity is outside a configurable range relative to the reference image, it's discarded; otherwise, it's added to the corresponding pixel's intensity in the first image. Finally, a display component produces an output image with the accumulated, stabilized pixel values.
16. The computer memories of claim 15 , wherein the first image represents a reference image, wherein the sequence of images further includes a third image, wherein the threshold component further compares intensity values associated with pixels in the third image to the intensity values associated with corresponding pixels in the reference image, and wherein the cumulative component further selectively adds the intensity values associated with the pixels in the third image to the cumulative intensity values based on the comparison.
The computer memories of claim 15 embodying computer-executable components, said components comprising: a memory component that when executed causes at least one processor to access a first image and a second image from a sequence of images captured by a computing device; a threshold component that when executed causes at least one processor to compare intensity values associated with pixels in the second image to intensity values associated with corresponding pixels in the first image; a cumulative component that when executed causes at least one processor to compensate for at least one of subject movement or movement of the computing device by selectively accumulating the intensity values associated with the pixels in the first image with the intensity values associated with the corresponding pixels in the second image based on the comparison performed by the threshold component to produce cumulative intensity values, the selective accumulating comprising, for each pixel in the second image, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to corresponding intensity value in the first image and producing a cumulative intensity value for that pixel; and a display component that when executed causes at least one processor to produce an output image having the selectively accumulated intensity values, where the first image acts as a reference image, adds a third image to the sequence and then the software compares pixel intensity values in the third image to the reference image as well, and selectively adds the intensity values from the third image to the total from the first and second images based on the same comparison process.
17. The computer memories of claim 15 , wherein the first image represents a reference image, wherein the threshold component further defines a range of expected intensity values for each of the pixels in the second image based on the intensity values associated with the pixels in the reference image.
The computer memories of claim 15 embodying computer-executable components, said components comprising: a memory component that when executed causes at least one processor to access a first image and a second image from a sequence of images captured by a computing device; a threshold component that when executed causes at least one processor to compare intensity values associated with pixels in the second image to intensity values associated with corresponding pixels in the first image; a cumulative component that when executed causes at least one processor to compensate for at least one of subject movement or movement of the computing device by selectively accumulating the intensity values associated with the pixels in the first image with the intensity values associated with the corresponding pixels in the second image based on the comparison performed by the threshold component to produce cumulative intensity values, the selective accumulating comprising, for each pixel in the second image, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to corresponding intensity value in the first image and producing a cumulative intensity value for that pixel; and a display component that when executed causes at least one processor to produce an output image having the selectively accumulated intensity values, where the first image is the reference image, defines an expected range of intensity values for each pixel in the second image, based on the intensity value of the corresponding pixel in the reference image.
18. The computer memories of claim 17 , wherein the threshold component compares the intensity values associated with the pixels in the first image to the intensity values associated with the corresponding pixels in the second image by comparing the intensity values associated with the pixels in the second image to the defined range of expected intensity values for each of the pixels in the second image.
The computer memories of claim 15 embodying computer-executable components, said components comprising: a memory component that when executed causes at least one processor to access a first image and a second image from a sequence of images captured by a computing device; a threshold component that when executed causes at least one processor to compare intensity values associated with pixels in the second image to intensity values associated with corresponding pixels in the first image; a cumulative component that when executed causes at least one processor to compensate for at least one of subject movement or movement of the computing device by selectively accumulating the intensity values associated with the pixels in the first image with the intensity values associated with the corresponding pixels in the second image based on the comparison performed by the threshold component to produce cumulative intensity values, the selective accumulating comprising, for each pixel in the second image, discarding an intensity value that is above a first threshold or below a second threshold, the first threshold and the second threshold being different from each other, otherwise adding the intensity value to corresponding intensity value in the first image and producing a cumulative intensity value for that pixel; and a display component that when executed causes at least one processor to produce an output image having the selectively accumulated intensity values, where the first image represents a reference image, wherein the threshold component further defines a range of expected intensity values for each of the pixels in the second image based on the intensity values associated with the pixels in the reference image, compares the pixel intensity values by checking if the intensity value of each pixel in the second image falls within its pre-defined, expected range derived from its corresponding pixel in the reference image.
19. The computer memories of claim 18 , wherein the cumulative component selectively accumulates the intensity values associated with the pixels in the first image with the intensity values associated with the corresponding pixels in the second image by discarding the intensity values associated with those pixels in the second image whose intensity values fall outside the defined range of expected intensity values.
This invention relates to image processing, specifically a method for combining intensity values from two images while filtering out unwanted data. The problem addressed is the need to accurately merge image data while excluding pixels that do not meet predefined intensity criteria, ensuring only valid data contributes to the final output. The system includes computer memories storing a first image and a second image, where each image comprises pixels with associated intensity values. A cumulative component processes these images by selectively accumulating intensity values from corresponding pixels in both images. The key innovation is the selective accumulation process, which discards intensity values from the second image if they fall outside a defined range of expected intensity values. This ensures only valid or relevant data is combined, improving the accuracy and reliability of the merged image. The defined range of expected intensity values acts as a filter, allowing the system to exclude outliers or noise that could distort the final result. The corresponding pixels in the first and second images are aligned spatially, ensuring proper alignment during accumulation. This method is particularly useful in applications requiring precise image fusion, such as medical imaging, surveillance, or scientific data analysis, where maintaining data integrity is critical. The selective accumulation mechanism enhances the robustness of the image processing pipeline by dynamically filtering invalid data.
20. The computer memories of claim 18 , wherein the cumulative component selectively accumulates the intensity values associated with the pixels in the first image with the intensity values associated with the corresponding pixels in the second image by adding the intensity values associated with the pixels in the first image with those corresponding pixels in the second image whose intensity values are within the defined range of expected intensity values to produce the cumulative intensity values.
This invention relates to image processing, specifically a method for combining intensity values from two images to produce a cumulative intensity representation. The problem addressed is the need to accurately merge intensity data from multiple images while filtering out values that fall outside a defined range of expected intensity values, ensuring only relevant data contributes to the final output. The system includes computer memories storing a first image and a second image, where each image comprises pixels with associated intensity values. A cumulative component selectively accumulates these intensity values by comparing the intensity values of corresponding pixels in the first and second images. Only those intensity values within a predefined range of expected intensity values are added together. This selective accumulation produces cumulative intensity values that represent a filtered combination of the two images, effectively suppressing noise or irrelevant data outside the expected range. The predefined range of expected intensity values is determined based on a statistical analysis of the intensity values in the first image, ensuring the accumulation process is adaptive to the input data. This approach enhances the accuracy and reliability of the cumulative intensity representation by excluding outliers or irrelevant variations. The method is particularly useful in applications requiring precise image fusion, such as medical imaging, remote sensing, or any scenario where multiple images of the same scene must be combined while minimizing noise or artifacts.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
April 10, 2013
June 27, 2017
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.